ChatGPT Visual Ads Move Into Image Generation in US Test
OpenAI will test ChatGPT visual ads during image generation in the US this month, backed by new measurement partners and brand-suitability pilots.
OpenAI will test ChatGPT visual ads inside its image generator in the US later in October 2026, starting with a small group of advertisers. The format puts picture-based creative in front of users while they are making an image. Brands can show what inspired a product, how it performs in use, or what a service makes possible. A limited set of ad buyers has already signed on.
The trial extends an ad format that has run in ChatGPT for seven months. Those earlier placements carried an advertiser's name, logo, headline, description, landing page and one image alongside a chat answer. Moving the inventory into image generation gives brands a larger, more visual slot at a point where the user is already scanning pictures.
Guardrails Around ChatGPT Visual Ads
OpenAI says every ad is labelled and kept apart from the picture the model generates, and that sponsors cannot shape the chatbot's replies or its output. The company excludes private chats and conversations it flags as emotionally vulnerable, and some ad categories are barred outright. Whether paying subscribers will encounter the new unit has not been disclosed.
Telling a sponsored image from a generated one is the first thing users will check, because the two sit in the same workspace during a single session.
Measurement and Brand Suitability
Attribution makes up the other half of the announcement. Conversion signals arrive through Hightouch, Tealium and LiveRamp, letting advertisers tie campaigns to sales instead of counting impressions alone. OpenAI is also testing brand-suitability checks with DoubleVerify and Integral Ad Science, the independent verification media buyers typically want before committing budget to a new placement.
OpenAI frames the format as a discovery channel where people learn what sets a brand apart, which puts it nearer to sponsored placement inside a creative tool than to a search result. The first advertiser cohort is small, and no pricing or firm launch date beyond the October window has been published.
| Element | Existing ChatGPT ad unit | New visual format |
|---|---|---|
| Placement | Card shown alongside answers | During image generation |
| Creative | Name, logo, headline, description, landing page, image | Images showing product inspiration, usage or experiences |
| Availability | 31 European countries, since August 2026 | US test starting late October 2026 |
| Verification | OpenAI's own reporting stack | Pilots with DoubleVerify and Integral Ad Science |
OpenAI puts the platform's weekly audience at 1.2 billion people. Its European ad inventory went live in August 2026 and now reaches 31 countries. Under the EU Digital Services Act, the company has to publish a complete public record of those ads by January 2027.
Why This Matters
Placing ads inside an image tool pushes labelling and separation rules in front of ordinary users, not regulators alone. Brands gain image inventory on a platform OpenAI says reaches 1.2 billion people each week, but the spending decision hinges on whether the measurement integrations and verification pilots satisfy buyers who want independent proof of placement. The European ad record due by January 2027 will be the first large public view of how the format behaves outside a controlled test.
Sources
Building advertising for the way people use AI
Photo by Zulfugar Karimov on Unsplash
Related Articles
- OpenAI Rewrites ChatGPT Advertising Policy to Block Rival Image and Audio AI Ads
- OpenAI's ChatGPT Sponsored Agents Turn Ad Clicks Into Brand Chats
- ChatGPT ads expansion reaches five new markets as OpenAI turns advertising into core business
✔Human Verified
Researched and cross-referenced against primary sources by the Bytevyte editorial team. This article was generated with the assistance of artificial intelligence and reviewed by the Bytevyte editorial team.